Summary Breakdown of the blood-retinal barrier (BRB) and thickening of the endothelial cell basement membrane due to deposition of extracellular matrix (ECM) are early events in the pathogenesis of diabetic retinopathy (DR). Current treatments are limited by significant side effects, including ECM deposition. Bone morphogenetic protein- 2 (BMP2), a secreted cytokine belonging to the TGF-? superfamily initiates signaling through activin receptors like kinases (Alk1, 2, 3, and 6) with high affinity to Alk2 and 3. The goal of this project is to test the hypothesis that in diabetes, BMP2 compromises BRB and induces ECM formation through the endothelial Alk2/3-dependent mechanism. Our hypothesis is supported by compelling preliminary findings that retinal and circulating BMP2 are upregulated in diabetic humans and mice, as well as in human retinal endothelial cells (HRECs) subjected to high glucose (HG). Importantly, BMP2 inhibitors attenuate the permeability and ECM deposition effect of HG in HRECs. Moreover, BMP2 activates the canonical and non-canonical pathways (smad/Runx2 and p38 MAPK/NF?B, respectively) in HRECs. Our hypothesis predicts that in diabetes BMP2 activates smad and p38 MAPK pathways integrate at wnt/?-catenin to induce hyperpermeability and ECM deposition. We will test the following specific aims: 1. Activated BMP2/Alks signaling system contributes to retinal endothelial cell dysfunction in DR: We will correlate changes in circulating levels of BMPs to the development of DR using serum samples from a large cohort of diabetic patients enrolled in the Phenome and Genome of Diabetes Autoimmunity (PAGODA) study and have been prospectively monitored for the development of diabetic complications including DR in the last 15 years. Using a streptozotocin-induced diabetic mouse model, and HRECs subjected to HG, we will determine the levels and distribution of BMP2 signaling system (BMP2, BMP4, BMP receptors, smad1/5/9, Runx2, and negative regulators of BMP2 such as noggin and BMP-binding endothelial regulator (BMPER). Subsequently, we will study the effects of pharmacological inhibition or genetic manipulation of the BMP2/Alks system on BRB function, and ECM. We will use endothelial Alk2/3 conditional knockout mice developed within our lab and commercially available BMPER-deficient mice (BMPER+/?) and; 2. Both canonical and non-canonical pathways contribute to BMP2-mediated retinal endothelial cell dysfunction in DR: We will test a) the effect of modulation of BMP2/Alks signaling on the smad1/5/9 and p38/NF?B pathways under normal or diabetic conditions, b) the effect of inhibition of smad1/5/9 or p38/NF?B signaling on diabetes or BMP2-induced permeability and ECM deposition, and c) the role of the wnt/??catenin pathway as a potential downstream target from both p38/NF?B and smad1/5/9 pathways to mediate retinal microvascular damage induced by BMP2. The translational significance of this proposal is the therapeutic potential of inhibition of BMP2/Alks signaling to improve the visual outcomes in DR with the ultimate goal of overcoming the limiting factors of current therapies in the prevention of ECM deposition.